LLM

Asking a model to "reply in JSON" works until it does not: a missing field, a trailing comment, a number returned as a string, a polite sentence before the open…

An agent that runs for four seconds inside one HTTP request can afford to fail: the user retries and nothing else happened. An agent that runs for eleven minute…

A base language model finishes text. It does not answer questions, refuse harmful requests, or stop talking at the right moment — those behaviors are installed …

A 70B-parameter model does not fit on an 80GB GPU, and the reason is not the parameters. In mixed-precision training with Adam you carry roughly sixteen bytes p…

An AI agent that calls tools sounds like it needs a framework. It doesn't. The model never runs anything itself: it returns a structured request to call a tool,…

A language model does not output text. At each step it outputs one score per vocabulary token, and a separate piece of code, the sampler, turns those scores int…

BM25 is a ranking function from the 1990s that still runs behind Lucene, Elasticsearch and OpenSearch, and it is still the baseline that new retrieval models ar…

Model APIs fail in ordinary ways: rate limits, overloaded capacity, server errors, dropped connections and calls that run longer than a timeout. Most of these a…

You have probably seen a claim shaped like this on a model card: "70B total parameters, but only 13B active per token." That is not marketing rounding — it desc…

A support ticket that needs docs, a tool call, and a model reply does not need twelve equal "frameworks." It needs an orchestration lane, a retrieval lane, a se…

A chatbot answers one prompt at a time. An agentic AI system accepts a goal, selects actions, calls tools, observes results, and loops until it reaches a termin…